Rameet Sangha and Andy Parkinson highlight developments in merger assessment across different sectors and jurisdictions, as well as issues raised by conglomerate mergers in a summary from the May 2022 Concurrences International Mergers conference.
Panel 1 – Latest developments in mergers assessment: what consequences on substance and procedure?
Rameet Sangha: As to whether merger control is stricter in some sectors than in others or between different jurisdictions, it can be seen that in non-technology cases the majority are approved in Phase I either without conditions or with conditions. Only 6% of cases are in Phase II. On the contrary, tech cases are less numerous. A smaller proportion of cases are approved in Phase I and a much higher proportion, about 24%, move to Phase II. It is, therefore, possible to see a real sectoral difference here. As in the EU, we see that in the technology sector in the UK there are relatively few cases compared to non-technology, but again, the most striking thing is that we have a much higher proportion going to phase II than for non-technology for the EC, and a significant proportion of these have been banned; there is a big dark red bar here. These findings raise questions about the reasons for these differences.
In the UK, when thinking about mergers, the focus has been on how companies compete, how competition works, and then how the merger will affect it.
As far as the type of evidence is concerned, if one looks at all of the European Commission’s merger decisions over time, the market share is quite explicit. This type of evidence is referenced in each of the decisions. Economic evidence may also be mentioned. This includes bidding studies, consumer surveys, the gross upward price pressure index, etc. Internal company documents are also explicitly used as evidence. A peak in 2016 can be seen, which reflects the Dow/Du Pont case, where there was a lot of reliance on this document, and it is a very long decision, which distorts the data. Within the types of evidence that the European Commission relies on, there are also the previous cases, the market investigations, which are increasingly used.
As far as the CMA’s practice is concerned, it is almost equivalent to that of the Commission. Indeed, there is the use of economic evidence, market shares are also mentioned, internal documents, and third-party evidence, which are similar to the European Commission’s market investigations.
Panel 2 – Conglomerate mergers: still waters run deep?
Andy Parkinson: Data collected since 1990 show that in Phase I decisions there has been a steady increase in the number of decisions per year that mention co-location. However, in Phase II, the number of investigations is much lower, but the number of cases mentioning bundling has increased since around 2010 to a slightly higher level, although it does not show the kind of steady growth that Phase I decisions show.
Concerning the mention of bundling in Commission decisions, the average mention will depend on whether a decision mentioned bundling at least once. In the context of Phase 1 decisions, a steady increase in the number of cases with many mentions of bundling can be observed. This indicates a more intensive examination of conglomerate effects in Phase I decisions, especially since 2015. As for Phase II decisions, there have been fewer. Thus, for example, in the General Electric/Honeywell decision there were 100 mentions, and then only in 2015 was there a more significant peak in the number of cases with mentions of hard selling and thus a very detailed examination of conglomerate effects in these cases.
The main concern from an economic point of view is that after the merger, a company can exercise market power from one market to an adjacent market, often by tying or bundling products together, and this can take various forms, including interoperability. In analysing such a merger, one must consider ability, incentive, and effect. First, concerning the ability to capture, a key question is whether the firm has significant market power. In particular, the product must be seen as sufficiently important by customers who therefore have few relevant alternatives. Secondly, concerning the type of bundling strategy that companies might adopt, one can broadly speak of mixed bundling, pure bundling or contractual tying, and technical tying. In this context, the economic literature points out that one should be less concerned about mixed bundling and more concerned about technical tying, pure bundling, and contractual tying. This requires very detailed work to examine how a company can effectively bundle or link its activities in practice. In a third step, the next key piece of information examined is the degree of common customers. This means that if two products are always used together, it will be possible to impose a flat rate on all sales. Finally, the effects are also examined. It is very important to note that a loss of sales from rivals is not a concern in itself. The concern is whether there are exclusionary effects, a reduction in the incentive, or the ability of rivals to compete. Thus, all of these points show that the assessment of conglomerate effects is complex and fact-based, and because there is little direct evidence, it is necessary to infer, based on market characteristics and counterfactual evidence, whether the market is compatible with conglomerate effects likely to occur.
This synthesis was originally published by Concurrences and is also available on their website (subscription required). The views expressed are those of the authors only and do not necessarily represent the views of Compass Lexecon, its management, its subsidiaries, its affiliates, its employees, or clients.